© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 43868
    Prev
    Next
    INDEX
    Explanations

    number of

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     somewhat
    -1.85
    
    -1.75
    多数
    -1.72
     wichtigsten
    -1.72
    ↵↵
    -1.71
    大多数
    -1.70
    
    -1.66
     Quando
    -1.58
    我们
    -1.55
     strikingly
    -1.55
    POSITIVE LOGITS
     which
    1.71
     kanilang
    1.70
     you
    1.63
    合せ
    1.59
     he
    1.57
     its
    1.56
    which
    1.55
    しい
    1.52
     ambayo
    1.52
    してしまった
    1.51
    Activations Density 0.042%

    No Known Activations